1. Field of the Invention
The present invention relates to computer graphics (CG) and, more specifically, to a technique for transforming a three-dimensional object in a virtual three-dimensional space.
2. Related Background Art
In CG, a technique called cluster transformation in general has been known as a technique for transforming a three-dimensional object whose form is determined by a plurality of vertexes, in order to express its fine actions and the like. In the cluster transformation, vertexes of a three-dimensional object are grouped into a plurality of groups, and each group forms a cluster. In each cluster, the vertexes belonging thereto are moved, so as to transform the three-dimensional object. Each vertex can belong to a plurality of clusters. Also, a weight can be set for the vertexes belonging to each cluster. Consequently, very fine transformations can be carried out as well.
The cluster transformation can effectively be utilized for transforming objects in video games as well. If the cluster transformation is employed for minutely deforming characters in a game, then it becomes easier to express them realistically.
Japanese Patent Application Laid-Open No. HEI 10-188028 has been known as an example in which the cluster transformation is applied to a video game. In the technique disclosed in this publication, for a three-dimensional object whose framework is defined by skeletons, joints and skeletons are also imparted to constituent parts lacking joints and bones in reality.
In the above-mentioned publication, a number of polygons constituting a three-dimensional object accompany a skeleton. A number of polygons accompanying a skeleton are controlled so as to rotate when the skeleton rotates. As a consequence, parts lacking joints and bones can be caused to perform minute movements. For example, in a three-dimensional object representing a human face, a polygon representing an eyelid is made to accompany a skeleton, and this skeleton is rotated. Hence, the three-dimensional object can be transformed such that the eyelid opens and closes.
However, a vertex rotating operation necessitates a greater use of matrix operations. Therefore, if there are a greater number of expression parts involving a minute transformation on a screen, then the load for processing them increases. Currently, there is a demand for a procedure which realizes a similar display mode while reducing the processing load.